Orthogonal integer transform method used for image and video compression

Abstract

An orthometric integer transforming method for the image and video compression, firstly dividing the image as 8*8 small blocks M, and processing the two-dimensional transformation for each small blocks M by the 8*8 orthometric transforming matrix for the compressing encoding to obtain the transforming coefficient matrix N and processing entropy encoding the coefficient quantified to complete the compressing process for the image and video; all the elements of the orthometric transforming matrix are all integers satisfying a certain condition; the element with different order or the elements with a different factors are considered as the same matrix. The invention designs a new 8*8 orthometric transforming matrix for forming a infinite integer matrix cluster; the invention generates the optimal integer transformation in the thousands of matrixes under the action of an optimal mathematic mode via the restriction of a certain condition to reach the theory optimization. The test indicates that the compressing function is fully greater than the DCT by the orthometric transforming compressed image and video.

Description

【Technical field】 [0001] The invention relates to the field of information technology, in particular to a method for digital image and digital video compression coding transformation. 【Background technique】 [0002] With the rapid development of technologies such as computers, microelectronics, information processing, communications, and lasers, multimedia technologies integrating graphics, text, sound, and images have rapidly penetrated into computers, communications, radio and television, and consumer entertainment industries. In the field of technology, more and more digital devices are adopted through digital signal transmission. Digital signals have many advantages, but when the analog signal is digitized, its frequency band will be greatly widened. For example, after a 6MHz ordinary TV signal is digitized, its digital rate will be as high as 167Mbps, which requires a lot of storage capacity and transmission bandwidth. Digital signals lose their practical value. Digit...

AI Technical Summary

Problems solved by technology

Therefore, a lot of progress has been made in integer transformation techniques and methods, but these integer transformations also have some shortcomings: the calculation performance of some integer transformations still has room for improvement, especially in the inverse transformation, the current calculation efficiency of integer transformations is not high

At present, the existing technical idea of ​​integer transformation is still to seek the approximate matrix of DCT, and the performance will decline to a certain extent after integer transformation

Images